Fujitsu is collaborating with BBVA, Spain’s second largest bank, on a proof of concept (PoC) of its quantum-inspired optimization solution Digital Annealer.
The bank wants to determine whether the Digital Annealer could outperform traditional computing techniques in optimising asset portfolios, helping minimize risk while maximizing returns, based on a decade’s worth of historical data.
BBVA is conducting multiple studies to determine how quantum technologies could represent an advantage over traditional tools in different financial use cases.
In support, BBVA launched a strategy that involves forging alliances, strengthening its internal capacity around this technology, evaluating different available tools, and developing proof of concepts in collaboration with research centers, traditional businesses and startups.
In addition to portfolio optimisation, BBVA researchers are also exploring how the technology can be applied to areas such as the simulation of financial scenarios, currency arbitrage and credit scoring processes.
Regularly optimising portfolios is a typical challenge for financial institutions. But while creating projections for individual assets can be relatively easily, balancing an entire investment portfolio is incredibly complex. This is because there are countless possible combinations. For example, finding the optimal selection from just 20 stocks generates the equivalent of over one quintillion (1018) permutations.
Because of this complexity, portfolio optimisation has traditionally been a manual task, guided more frequently by guesswork than empirical data – simply because the convoluted calculations far exceed the capabilities of regular computers.
However, Fujitsu’s Digital Annealer has been designed to process exactly this sort of complex combinatorial problem in just minutes.
BBVA’s Research and Patents team wanted to explore how Digital Annealer could be deployed to optimise an investment portfolio by evaluating different assets based on information including historical market prices to determine the level of investment in each.
The bank also intends to use Fujitsu Digital Annealer to identify the best possible time to buy or sell assets, to maximise portfolio value.
The project is still in an exploratory phase, but early results point to a set of advantages for the Digital Annealer in comparison to tools currently used to resolve complex problems quickly, accurately, and efficiently. Specifically, the PoC between BBVA and Fujitsu has determined that better results can be obtained for large-scale calculations involving more than 100 assets or factors.
Carlos Cordero, chief technology officer at Fujitsu in Spain, comments: “All industries have multiple areas that would benefit from optimization however generally the sheer complexity of the problems means that guesswork or trial and error are the most common methods.
“While true quantum computing as a technology is still in the laboratory testing phase, Digital Annealer represents a bridge to this future technology, thanks to its ability to evaluate multiple different combinations extremely rapidly. Bringing this optimisation to the financial industry, for example to identify the optimal portfolio and the ideal time to acquire or offload assets can make a huge difference to the returns it delivers.”
Carlos Kuchkovky, BBVA global head of research and patents, explains: “The quantum technologies ecosystem is evolving very quickly and we believe that the collaboration with various partners, public and private alike, is key in order to translate the benefits of the technology into tangible progress, for the banking sector and for society at large. Although this technology is still in an early stage of development, its potential to impact the sector is already a reality.
“Our research is helping us identify the areas where quantum computing could represent a greater competitive advantage, once the tools have sufficiently matured. At BBVA, we believe that quantum technology will be key to solving some of the major challenges facing society this decade.
“Addressing these challenges dovetails with BBVA’s strategic priorities, such as fostering the more efficient use of increasingly greater volumes of data for better decision-making as well as supporting the transition to a more sustainable future.”
Fujitsu has already deployed the Digital Annealer to solve multiple different problems for many industries, including identifying the optimal seam patterns for welding robots to follow on an automotive assembly line, determining the best routes for delivery trucks, finding the ideal parts placement and picking routes for distribution warehouses, maximizing return on investment for utilities companies and, in the pharmaceutical sector, accelerating the discovery of new substances and development of new drugs.